This dissertation proposes a uniqueness-based theory of definiteness that accounts for the semantics and the pragmatics of German (non-) contracted forms of a preposition and the definite determiner (e.g. ‘zum / zu dem’, ‘am / an dem’), but the analysis also carries over to definite descriptions in general. I propose that the definite determiner is ambiguous between a referential and a quantificational reading. Crucially, referential DDs exploit information provided in the surrounding linguistic context, while the interpretation of quantificational DDs heavily relies on extra-linguistic world knowledge that can be represented by an implicit free individual variable and an implicit free relation variable. Non-contracted forms are always interpreted referentially, whereas contracted forms receive quantificational interpretations. This proposal has a wide range of applications: It deals with anaphoric and demonstrative DDs, as well as with typical ‘uniqueness uses’ (such as ‘the moon’ or ‘the sun’), covarying DDs involving explicit and implicit antecedents, bridging definites, and, last but not least, so-called Weak Definites.